{"data":{"sci_version": "1.0", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:40113641", "kind": "science-record", "revision": 2, "content_hash": "sha256:2ee2866c79b19806dda1e6b1bf233f405bc2ad0439f30bf1ab2fe8ca58bc090e", "created_at": "2026-11-01T10:25:50Z", "updated_at": "2026-11-01T14:01:38Z", "generator": {"name": "corpus-assay-engine", "converter": "sep_to_sci_v1", "source_package_schema": "urn:bntrg:sep:1.0.0"}, "paper": {"state": "partial", "title": "Expression and prognosis of NR3C1 in uterine corpus endometrial carcinoma based on multiple datasets.", "doi": "10.1007/s12672-025-02086-1", "pmid": "40113641", "authors": null, "journal": "Discover oncology", "year": 2025, "abstract": null, "claims": [{"id": "claim-1", "statement": "the headline numerical result the paper states for its main finding", "kind": "quantitative", "state": "complete", "target": {"value": 2.17, "unit": "hazard ratio", "tolerance": null, "source": "chars 12212-12253 of source sha256 a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c"}, "evidence": [{"asset": "source:pmid:40113641", "locator": "chars 12212-12253 of source sha256 a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "quote": "NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***"}]}], "gaps": ["abstract", "authors"]}, "experiment": {"state": "partial", "objective": "the headline numerical result the paper states for its main finding", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate and its interval from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The number a replication must land on, and the sentence it came from.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:40113641"], "outputs": ["the published estimate, its unit and its interval"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data behind the published result", "kind": "data-preparation", "state": "blocked", "blocked_on": "data:not_downloaded", "description": "The published result cannot be recomputed without the data the paper analysed.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the source data with its provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the analysis programme", "kind": "analysis", "state": "blocked", "blocked_on": "no owned computational capability is bound to this experiment", "description": "Re-run the paper's analysis on the source data and produce the registered result. | SEP input not carried into the asset list: materials:the source data behind the published result in pmid:40113641", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the registered result"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40113641' does not resolve to a declared asset"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:40113641", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.1007/s12672-025-02086-1", "sha256": "sha256:a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.1007/s12672-025-02086-1", "version": null, "note": "Source read by the package builder: full_text.", "structure": {"access_state": "full_text"}}, {"id": "asset:table:pmid:40113641:1", "role": "table", "format": "jats-table", "media_type": "application/vnd.jats.table+json", "uri": null, "sha256": "sha256:68444c5cd906dee9259abd59486ecb0e15e60e38cddf62ad685b5ea579d66c90", "bytes": 645, "access": "open", "state": "verified", "reader": "read_table", "reader_status": "available", "identifier": "section[19]/table-wrap[1]#Tab1", "version": null, "note": "Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysis", "structure": {"pmcid": "PMC11926321", "section_index": 19, "section_title": "Cox risk regression analysis in UCEC", "row_count": 12, "column_count": 3}, "table": {"label": "Table 1", "caption": "Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysis", "rows": [["Clinical characteristics", "HR (95% CI)", "p-value"], ["Age", "", ""], ["≤ 60", "Ref", ""], ["> 60", "2.94 [1.14–2.89]", "0.012*"], ["Stage", "", ""], ["Stage I", "Ref", ""], ["Stage II", "2.62 [0.82–3.59]", "0.154"], ["Stage III", "4.55 [1.96–5.17]", "< 0.001***"], ["Stage IV", "12.39 [4.69–15.39]", "< 0.001***"], ["NR3C1", "", ""], ["Low", "Ref", ""], ["High", "2.17 [1.29–3.04]", "0.00173**"]], "row_count": 12, "column_count": 3, "numeric_cells": 13, "header_rows": 1, "locator": "section[19]/table-wrap[1]#Tab1"}}, {"id": "asset:table:pmid:40113641:2", "role": "table", "format": "jats-table", "media_type": "application/vnd.jats.table+json", "uri": null, "sha256": "sha256:0d73e56ab69dda9a9eb4c956d8286873744af7815ccac5a911cfbb3be1111b9c", "bytes": 1667, "access": "open", "state": "verified", "reader": "read_table", "reader_status": "available", "identifier": "section[23]/table-wrap[1]#Tab2", "version": null, "note": "Gene sets enrichment from GSEA of NR3C1 in UCEC", "structure": {"pmcid": "PMC11926321", "section_index": 23, "section_title": "Gene sets enriched in NR3C1 expression phenotype", "row_count": 25, "column_count": 5}, "table": {"label": "Table 2", "caption": "Gene sets enrichment from GSEA of NR3C1 in UCEC", "rows": [["Gene set name", "Set size", "NES", "NOM p-val", "FDR Q-val"], ["KEGG", "", "", "", ""], ["KEGG_RIBOSOME", "151", "− 2.734", "0.017", "0.035"], ["KEGG_OXIDATIVE_PHOSPHORYLATION", "106", "− 1.907", "0.011", "0.027"], ["KEGG_TH1_AND_TH2_CELL_DIFFERENTIATION", "90", "1.830", "0.001", "0.005"], ["KEGG_TH17_CELL_DIFFERENTIATION", "106", "1.817", "0.001", "0.005"], ["KEGG_SYSTEMIC_LUPUS_ERYTHEMATOSUS", "51", "1.813", "0.001", "0.005"], ["GO_BP", "", "", "", ""], ["GO_IMMUNOGLOBULIN_PRODUCTION", "199", "1.984", "0.001", "0.002"], ["GO_PRODUCTION_OF_MOLECULAR_MEDIATOR_OF_IMMUNE_RESPONSE", "313", "1.885", "0.001", "0.002"], ["GO_ CELL_MATRIX_ADHESION", "235", "1.746", "0.001", "0.002"], ["GO_ANTIGEN_RECEPTOR_MEDIATED_SIGNALING_PATHWAY", "194", "1.735", "0.001", "0.002"], ["GO_NEGATIVE_REGULATION_OF_BLOOD_VESSEL_MORPHOGENESIS", "140", "1.725", "0.001", "0.002"], ["GO_CC", "", "", "", ""], ["GO_EXTERNAL_SIDE_OF_PLASMA_MEMBRANE", "408", "1.818", "0.001", "0.004"], ["GO_T_CELL_RECEPTOR_COMPLEX", "131", "1.761", "0.001", "0.004"], ["GO_BLOOD_MICROPARTICLE", "142", "1.755", "0.001", "0.004"], ["GO_PLASMA_MEMBRANE_SIGNALING_RECEPTOR_COMPLEX", "302", "1.645", "0.001", "0.004"], ["GO_SARCOLEMMA", "133", "1.636", "0.001", "0.004"], ["GO_MF", "", "", "", ""], ["GO_ANTIGEN_BINDING", "115", "2.101", "0.001", "0.005"], ["GO_EXTRACELLULAR_MATRIX_STRUCTURAL_CONSTITUENT", "164", "1.829", "0.001", "0.005"], ["GO_CYTOKINE_BINDING", "128", "1.755", "0.001", "0.005"], ["GO_INTEGRIN_BINDING", "144", "1.622", "0.001", "0.005"], ["GO_IMMUNE_RECEPTOR_ACTIVITY", "136", "1.608", "0.001", "0.005"]], "row_count": 25, "column_count": 5, "numeric_cells": 81, "header_rows": 1, "locator": "section[23]/table-wrap[1]#Tab2"}}], "gaps": ["asset:source:pmid:40113641 not verified"]}, "analysis": {"state": "absent", "methods": [], "targets": [], "gaps": ["method", "targets"]}, "capability": {"state": "absent", "requirements": [], "gaps": ["capability"]}, "readiness": {"state": "partial", "tier": null, "blocking_axis": null, "blocked_on": null, "axes": {}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:40113641", "asset:table:pmid:40113641:1", "asset:table:pmid:40113641:2"], "steps": ["read_published_estimate"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": ["xml"], "capabilities": [], "data_gaps": ["source:pmid:40113641"]}, "gaps": ["corpus_record_absent"]}, "review": {"state": "partial", "machine_critique": {"status": "blocked", "providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "full-text-reading-v1", "prompt_hash": "9763abbafa37b97b90da11b8d6525a9a84a464c0f42eebd6f59086448a9c1a8f", "source_hash": "a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "raw_response_sha256": "sha256:1d5d2b1621eb12f9b6b002fd93cec3eb4fe86b90f7cecb6f0104a476e99d9e64", "raw_response_chars": 4979}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "full-text-reading-v1", "prompt_hash": "84b221a0636c32dd2befb500f82e712f7cba8efe5d103f2b087f3c8b3a3fdecd", "source_hash": "a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "raw_response_sha256": "sha256:1923651e9a2986f7e719c86ca9d21f9385359b45f0b36e32ee288e3c33a43f49", "raw_response_chars": 61641}], "target_providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "replication-target-v2", "prompt_hash": "7487cd327d624983212c72a9c9d83b90fe1a5c796138f1f4b57f6f9c4daff48b", "source_hash": "a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "raw_response": "{\"parent_value\":2.17,\"unit\":\"HR\",\"evidence\":\"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\",\"target_result\":\"High NR3C1 expression was associated with a hazard ratio of 2.17 for UCEC overall survival.\",\"number_stated\":true}", "reading": {"parent_value": 2.17, "unit": "HR", "evidence": "NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***", "target_result": "High NR3C1 expression was associated with a hazard ratio of 2.17 for UCEC overall survival.", "number_stated": true, "source_locator": "chars 12212-12253 of source sha256 a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c"}, "recorded_at": "2026-10-31T15:36:08.259239+00:00"}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "replication-target-v2", "prompt_hash": "19e3d6df97836489f339d64d72aea703bdca3b59a34972e7cc6dcba7daae2da2", "source_hash": "a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c", "raw_response": "<think>We need answer JSON only. Need identify single numerical result replication registered target headline numerical result paper states for main finding. Need interpret likely target perhaps NR3C1 main finding headline numerical result. Source paper abstract-like text has many numbers. User says target \"the headline numerical result the paper itself states for the registered target\" vague. Need find a single numerical result that replication of registered target has reproduce. Could be perhaps AUC for NR3C1, likely main finding. Need determine which headline number? Let's parse.\n\nTitle Expression and prognosis of NR3C1 in UCEC. Main finding NR3C1 potential biomarker diagnosis/prognosis. ROC across four datasets AUC 0.914, 0.791, 0.786, 0.882. Cox HR high NR3C1 2.17 [1.29–3.04], p=.00173. Which is \"headline numerical result\" likely target registered perhaps benchmark extraction task where registered target is main result, and ask number paper states. Could be \"area under curve 0.914\" likely headline. Need not calculate. They demand single numerical result and exact contiguous quote. The registered target description generic because perhaps expected identify paper's headline number. Maybe 72 co-expressed genes? Abstract says analysis yielded 72 co-expressed genes; top 10; NR3C1 impact; maybe headline numerical result is 72. The replication of registered target \"the headline numerical result the paper itself states for its main finding\" likely target could refer to article title/main entity NR3C1, and numeric result perhaps hazard ratio? Let's inspect wording carefully.\n\nPrompt: \"Read supplied paper text and find the single numerical result that a replication of the registered target has to reproduce. Report the headline number only: the value the paper itself states for the registered target, not a value you calculated, converted, averaged or reconstructed. The unit must be the unit the paper prints. ... REGISTERED TARGET: the headline numerical result the paper itself states for its main finding\"\n\nThis appears generalized extraction. Registered target is \"the headline numerical result the paper itself states for its main finding\" rather than NR3C1? Maybe expected identify headline numerical result in source. \"parent_value\" perhaps benchmark field naming number. Need output target_result likely description. Need exact quote.\n\nCould headline numerical result be cohort/sample size 516 UCEC samples? Or 72 genes? Let's understand source likely paper with figures, results sections. Main finding is NR3C1 diagnostic and prognostic. Headline number likely AUC perhaps 0.914 from GSE17025. But there are multiple AUCs. The instruction \"single numerical result\" perhaps they expect one registered target specified elsewhere but here target phrased generically. Maybe title/abstract contains no AUC in abstract summary except p&lt;0.05. Main result in body. \"the value paper itself states for the registered target\" could refer to target itself being NR3C1. Numerical results attached to NR3C1 include AUCs, cutoff, HR, correlations. Which is \"headline\"? The most headline likely HR = 2.17 for high NR3C1, because prognosis is main finding, while diagnosis AUCs are multiple datasets.\n\nMaybe \"parent_value\" schema hints benchmark asks parent value of a table? But registered target text custom. Let's not assume.\n\nLet's analyze article structure and likely paper metadata. Maybe this source is an article where specific \"registered target\" is an experimental target like NR3C1, and task asks numerical result to reproduce. They might have registered \"NR3C1\" as target? The target line says \"the headline numerical result the paper itself states for its main finding,\" not \"NR3C1.\" It asks find \"single numerical result that a replication of the registered target has to reproduce.\" Perhaps the registered target could be \"headline numerical result\" not a biological target. Need return one.\n\nCould be from \"PaperQA / replication target\" tasks where each source has a \"registered target\" perhaps a numerical result. They tell us target in natural language. Need extract exact result.\n\nLet's inspect supplied text for likely headline. First summary paragraph:\n- 30% increase prevalence context\n- 4 datasets\n- 72 co-expressed genes\n- top 10 genes\n- NR3C1 impact\n- clinical factors\n- top 50 positively and negatively correlated genes\nThis abstract doesn't have performance number besides counts.\nThen sections:\n- enrichment 72 genes\n- ROC four datasets: AUC values\n- Cox 516 samples and HR\n- immune 22 types etc.\nMain finding in title: \"Expression and prognosis of NR3C1...\" A replication of NR3C1 finding perhaps HR 2.17 is the key prognostic result. The diagnostic ROC AUC also key.\n\nQuestion wording \"headline number only: the value the paper itself states for the registered target\" Perhaps target result is explicitly registered as some numerical target. Could \"main finding\" be the NR3C1 high expression hazard ratio 2.17. The exact quote could be table row:\n\"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nWould need quote whole clause if linked via \"respectively\"/parenthetical. But evidence exact contiguous quote containing number. Could quote a larger contiguous section. Unit likely \"HR\" (hazard ratio). parent_value 2.17. target_result e.g. \"High NR3C1 expression had a hazard ratio of 2.17 for overall survival in UCEC.\" number_stated true.\n\nOr headline AUC maybe 0.914 (unit \"AUC\"). Evidence exact:\n\"The ROC curve of GSE17025 has an offline area of 0.914 and a sensitivity of 81 30%, specificity 100.00%, Cut-off value 2305.500, indicating that in this dataset, NR3C1 gene expression levels above 2305.500 can be classified as UCEC, while expression levels below 2305.500 can be classified as non UCEC.\"\nThis attaches AUC to dataset, not necessarily headline.\n\nCould headline numerical result be AUC values all four exceeded 0.78. \"the AUC values for all four datasets exceeded 0.78\" is a single headline threshold? Replication target might reproduce \"AUC >0.78\" overall. parent_value 0.78, unit \"AUC\". Evidence exact clause. But user asks \"single numerical result\" and value paper states. 0.78 is a summary, but not exact measured AUC for target, and \"exceeded\" not equal. Probably not.\n\nMaybe \"single numerical result\" is \"516 UCEC samples\" because replication of Cox analysis must reproduce dataset. No.\n\nLet's seek cues in phrase \"parent_value\". Maybe generated schema generic for \"target result\" tasks. It could refer to \"parent_value\" because a numerical result might be parent metric in a figure (e.g., overall survival rate). They expect numeric value and unit. Maybe target_result free string.\n\nWe need only JSON. We need choose likely answer. Let's deeply reason what \"headline numerical result\" usually means in scientific paper. Abstract's key quantitative result often sample size or effect size. Here abstract summary omits numeric effect except p<0.05. The most prominent quantitative finding from title and main result is NR3C1 identified via survival regression. Table gives HR 2.17. Since high vs low expression independently influences overall survival, HR 2.17 is perhaps headline numeric effect. The ROC results (AUC) support diagnosis, but there are four. The question asks \"single numerical result\" likely wants one. If target were \"headline numerical result paper states for its main finding,\" the answer could be 2.17. Unit must unit paper prints: \"HR (95% CI)\" perhaps unit \"HR\". Evidence should quote table row maybe contiguous. Could quote prose plus table? Exact contiguous source includes concatenation:\n\"Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]&lt; 0.001***Stage IV12.39 [4.69–15.39]&lt; 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***P &lt; 0.05, **P &lt; 0.01, ***P &lt; 0.001\"\nCould quote whole table clause to establish row attribution. But evidence exact contiguous quotation and target linking. \"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\" is exact, though formatting collapsed. It contains number. Unit \"HR\". target_result say \"Overall-survival hazard ratio for high versus low NR3C1 expression: 2.17.\" Does unit include \"HR (95% CI)\"? They say unit paper prints. \"HR\" is dimensionless. Could say \"HR\". Number only in parent_value.\n\nBut maybe expected AUC 0.914. Which is more \"headline\"? Let's identify wording \"the headline numerical result the paper itself states for its main finding.\" In main conclusion, no numbers. Main finding title \"Expression and prognosis of NR3C1 in uterine corpus endometrial carcinoma based on multiple datasets\". Often replication target would be a specific numeric finding: e.g., \"NR3C1 diagnostic performance\" AUC. A replication must reproduce AUC values perhaps primary dataset 0.914. The paper's \"headline\" might be that NR3C1 is downregulated and high expression predicts worse prognosis; HR 2.17 is headline numerical result. I'd lean HR.\n\nCould main target instead be \"72 co-expressed genes\" from multiple datasets, because that's first key numerical result and abstract's headline. The registered target wording perhaps not biological target but \"headline numerical result.\" The paper's abstract explicitly says \"Analysis of four UCEC databases ... yielded 72 co-expressed genes.\" That's likely the headline numeric count. A replication of main finding would reproduce 72 genes. Yet target line \"for the registered target\" maybe target = paper's main finding, and number is 72. But what is \"unit\"? \"co-expressed genes\". They ask unit paper prints. Could output 72, unit \"co-expressed genes\", evidence exact quote. But this seems less likely because main finding is NR3C1, not number of genes.\n\nLet's parse sentence semantics of prompt generated:\n\"Read the supplied paper text and find the single numerical result that a replication of the registered target has to reproduce. Report the headline number only: the value the paper itself states for the registered target, not a value you calculated, converted, averaged or reconstructed.\"\n\n\"registered target\" may be an entity (e.g., \"the headline numerical result the paper itself states for its main finding\")—odd because target itself is described as numerical result. Maybe target registry could register a result like \"Overall survival hazard ratio for NR3C1\" but user only supplied broad descriptor due privacy? Yet says REGISTERED TARGET: the headline numerical result the paper itself states for its main finding. So target is literally conceptual.\n\n\"Attribution may be carried by the text's own wording: a number may be attached to the target by a parenthetical list, by the word 'respectively', or by words that name the target directly. When that is how the text links them, quote the whole clause so a reader can check the attribution for themselves.\"\n\nThis generic instruction targets extraction where numerical value might be one among several, need quote clause. Here target may be NR3C1. Numbers directly named near NR3C1. \"registered target\" probably asks find numerical result associated with NR3C1. We need infer headline.\n\nPotential numerical results directly tied to NR3C1:\n- ROC AUC each dataset: 0.914, 0.791, 0.786, 0.882\n- cutoffs 2305.500, 2.926, 8.750, 6.636\n- sensitivity/specificity\n- Cox HR 2.17\n- correlations with immune cells.\n\"Headline\" likely the one central metric, perhaps AUC. If article title includes \"Expression and prognosis\", prognosis's HR is central. The Cox section perhaps results heading and Fig. 6 main finding. It says NR3C1 expression as independent prognostic factor. HR 2.17 is the single effect estimate. ROC has 4, so not single. Thus likely 2.17.\n\nNeed exact evidence quote. Could quote from Table 1 with headers and row. The exact source includes weird no whitespace. We can quote:\n\"Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]&lt; 0.001***Stage IV12.39 [4.69–15.39]&lt; 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***P &lt; 0.05, **P &lt; 0.01, ***P &lt; 0.001\"\nLong but guarantees attribution. Or quote just:\n\"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nThis is contiguous and row ordering indicates Low Ref, High 2.17. A reader can check. But instruction \"When that is how text links them, quote whole clause so reader can check attribution.\" Table row itself enough. They demand evidence exact contiguous quotation containing number; no paraphrase. We could quote full table header + row, contiguous. JSON escape etc.\n\nCould unit be \"HR (95% CI)\"? The numeric is HR, confidence interval. \"The unit must be the unit the paper prints.\" Since table header prints \"HR (95% CI)\", perhaps unit should be \"HR\". Usually answer unit perhaps \"hazard ratio\". But exact printed unit is \"HR\". Use \"HR (95% CI)\" perhaps parent value 2.17 and unit \"HR (95% CI)\". Yet evidence row includes 2.17 [1.29–3.04]. They say headline number only. Unit should be the unit paper prints: likely \"HR\". If target_result mentions 95% CI maybe then number only conflict? Better unit \"HR\".\n\nLet's not rush. Need ensure no alternate likely expected \"0.914\" due ROC diagnostic main result. Search article likely original title. Maybe abstract omitted specific numbers because supplied text begins a summary generated? Source includes article title and sections. Could be paper in Journal of Ovarian Research? DOI s12672 is Journal of Ovarian Research (2025). Maybe full article. Search mentally: \"Expression and prognosis of NR3C1 in uterine corpus endometrial carcinoma based on multiple datasets\" likely abstract:\nBackground...\nMethods...\nResults: \"Analysis of four UCEC datasets yielded 72 co-expressed genes. ... Survival regression analysis highlighted NR3C1...\" \nConclusions.\nNo quantitative abstract results. The figures show:\nFig 2 enrichment\nFig 3 expression? \nFig 4 PPI?\nFig 5 ROC\nFig 6 Cox\nFig 7 immune\nFig 8? GSEA\nFig 9 LinkedOmics\nThe \"headline numerical result\" could refer to ROC AUC in Fig 5. In replication, for target NR3C1, perhaps need reproduce \"AUC of 0.914\" from GSE17025 as main diagnostic result. But why choose first of four? The summary says AUC values all > .78, no one headline. Maybe target could be \"NR3C1\" and replication requires all? Yet asks single numerical result, likely one explicitly singled by paper as strongest: 0.914 in GSE17025. It is highest AUC. But instruction says not value calculated or averaged; paper states it. Could be that.\n\nMaybe \"registered target\" comes from some benchmark where users register a \"replication target\" as a result phrase. For example they might ask:\nREGISTERED TARGET: the headline numerical result the paper itself states for its main finding\nThis is a generic target, and likely one number intended by source's title/abstract. The main finding maybe \"NR3C1 high expression HR = 2.17\". The highest AUC is not main finding per se, diagnostic subfinding. Prognostic HR is closer title.\n\nLet's consider abstract's wording: \"Survival regression analysis highlighted NR3C1 as the gene with a substantial impact on UCEC prognostic outcomes.\" Cox regression identified factors. The headline is NR3C1 itself, but number from Table 1 is effect. If asked to report \"headline number only,\" likely answer HR 2.17. They likely expect p-value perhaps 0.00173? In replication, headline result of prognostic factor is usually hazard ratio, with p-value support. \"single numerical result\" means perhaps hazard ratio. Unit HR. Yes.\n\nCould output target_result: \"High versus low NR3C1 expression was associated with a 2.17 hazard ratio for overall survival in UCEC.\" Evidence quote table row. number_stated true.\n\nNeed exact evidence no ellipsis. We can include exact sentence before and table:\n\"The analysis outcomes affirmed a significant correlation between clinical stage III, clinical stage IV, age, and the expression of NR3C1 with the occurrence and progression of UCEC (p &lt; 0.05) (Fig. 6a–c) (Table 1). These findings further confirm that clinical stage III, clinical stage IV, age, and NR3C1 are independent factors influencing the prognosis of UCEC.\"\nBut that doesn't include 2.17. Need contiguous quote reaching table row, could start at \"Fig. 6a–c...\" and include table. Let's ensure exact punctuation/spacing. Source:\n\"Fig. 6a–c Cox analysis of clinical stage, age, NR3C1 expressionTable 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref&gt; 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]&lt; 0.001***Stage IV12.39 [4.69–15.39]&lt; 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***P &lt; 0.05, **P &lt; 0.01, ***P &lt; 0.001\"\nCould quote from \"Table 1Association...\" to row. Exact. JSON handles &lt; as normal. This is long but okay.\n\nCould quote only exact contiguous:\n\"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nThis is perhaps not a clause but table row. Attribution clear. Requirement \"do not shorten it\" means do not shorten evidence quote? They say \"The evidence must be an exact, contiguous quotation from SOURCE TEXT that contains the number; do not shorten it, do not add an ellipsis and do not paraphrase it.\" This likely means quote enough/full sentence, not manipulate. \"do not shorten it\" could mean must quote complete relevant clause, perhaps table row is full row. We can include header+row to avoid ambiguity.\n\nWhat exactly target_result should say if true? They ask report headline number only but schema includes target_result. Could say \"The overall-survival hazard ratio for high NR3C1 expression versus low expression was 2.17.\" Is that \"number only\" no, but target_result likely needs description. Evidence quote. Maybe \"High NR3C1 expression: HR 2.17 (95% CI 1.29–3.04), p = 0.00173.\" But they explicitly report headline number only, so parent_value 2.17, unit HR; target_result can mention comparison. Avoid including CI/p in target_result maybe not headline. Evidence necessarily includes.\n\nMaybe answer:\n{\n \"parent_value\":2.17,\n \"unit\":\"HR\",\n \"evidence\":\"Table 1Association ... NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\",\n \"target_result\":\"The Cox overall-survival hazard ratio for high NR3C1 expression versus low NR3C1 expression in UCEC was 2.17.\",\n \"number_stated\":true\n}\nNeed exact evidence includes table header and all between. Let's ensure no accidental non-exact formatting. We can quote from source exactly:\n\"Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]&lt; 0.001***Stage IV12.39 [4.69–15.39]&lt; 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nThis is exact contiguous, yes. Unit \"HR (95% CI)\" maybe header. The result value 2.17 is a hazard ratio, with CI. Since \"unit\" should unit paper prints, perhaps \"HR (95% CI)\" isn't a unit but table heading. Use \"HR\".\n\nHowever, let's challenge whether perhaps expected 72. How to infer from \"headline numerical result the paper itself states for its main finding\". If someone asks paper extraction and target generic, what would they call headline numerical result? In article summary, first sentence says 30% increase, but that's background not result. Abstract's only numeric result is \"72 co-expressed genes\". Also \"top 10 genes\" and \"top 50...\" but those are methods/results. The headline numerical result might thus be 72. Yet main finding is NR3C1; the paper doesn't state a headline numeric effect in abstract. They might want 72 because that's prominent. But unit \"co-expressed genes\". Replication of \"main finding\" perhaps number of genes to identify. The phrase \"for the registered target\" suggests target could be \"NR3C1\" (main gene), and headline number associated with it is HR 2.17. Let's map:\nREGISTERED TARGET: the headline numerical result the paper itself states for its main finding\nThis target is not \"NR3C1\"; it's literally a placeholder description. Maybe the benchmark's registered target could be an entire paper result and they want identify numerical output. They use \"parent_value\" perhaps output is extracted from source. The intended one likely based on title/abstract. They ask \"a replication of the registered target has to reproduce\" meaning if registered target is e.g. \"the headline numerical result\", replication must reproduce that number. We should choose result central to paper.\n\nTitle includes \"Expression and prognosis of NR3C1\". Headline numerical result can be:\n- expression differential maybe log2 fold changes, not supplied in text? no.\n- prognosis: HR 2.17.\nThus HR is indeed central and single.\n\nLet's inspect Cox table: NR3C1 high vs low HR 2.17. This is likely main inferential statistic. The paper explicitly identifies independent prognostic factor. So use that.\n\nCould unit be \"hazard ratio (95% CI)\" to match printed header. parent_value 2.17. target_result perhaps \"Overall survival HR for high vs low NR3C1 expression: 2.17.\" Good.\n\nNeed JSON object no markdown. Ensure exact quote perhaps only table row. \"do not shorten it\" - include full clause. A table row is not clause but enough. They specifically say when attached by wording, quote whole clause. Here target is directly named in table row. Quote row:\n\"NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nThis is exact contiguous. But readability poor due source extraction. Add header contiguous. Better.\n\nCould include entire table from header through footnote; this demonstrates unit and result. But evidence should not be needlessly huge. Yet exact and clear. We can quote:\n\"Clinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]< 0.001***Stage IV12.39 [4.69–15.39]< 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nThis starts after title but includes headers. Is \"Clinical characteristics...\" exact contiguous yes. Better perhaps table title through row. No issue.\n\nLet's think whether evidence may need include prose establishing overall survival. Table title is before headers. Include it:\n\"Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]< 0.001***Stage IV12.39 [4.69–15.39]< 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\"\nExactly.\n\nCould there be issue unit \"HR (95% CI)\" because table prints it. But unit field should string. I'd use \"HR\" because number is point estimate. The printed unit is HR. If they expected \"hazard ratio\", likely acceptable semantically but exact \"HR\" safest.\n\nNeed only JSON. Before final, let's explore if any hidden issue: They ask \"Report the headline number only: the value the paper itself states for the registered target, not a value you calculated, converted, averaged or reconstructed.\" If number_stated false only if no number. We have true. \"single numerical result\" parent_value numeric. \"unit must be unit paper prints.\" HR. \"evidence exact contiguous quotation\" yes. \"Attribution may be carried by text's own wording...\" table direct. \"When that is how text links them, quote the whole clause...\" We quote table title/header/row. Great.\n\nPotential target result could mention \"overall survival\" as paper table title. Good.\n\nLet's consider p=0.00173 vs HR. They might view \"headline number\" as p-value because text says \"significant correlation ... (p < 0.05)\" but table exact p is 0.00173. Yet replication target main finding likely effect size HR 2.17, not p. We should not report p. Our target_result explicitly frames HR.\n\nMaybe \"headline numerical result\" in context of registered target could be \"NR3C1 high expression HR\" indeed. Good.\n\nLet's final JSON with no markdown.</think>\n\n{\"parent_value\":2.17,\"unit\":\"HR\",\"evidence\":\"Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]< 0.001***Stage IV12.39 [4.69–15.39]< 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***\",\"target_result\":\"The overall-survival hazard ratio for high versus low NR3C1 expression in UCEC was 2.17.\",\"number_stated\":true}", "reading": {"parent_value": 2.17, "unit": "HR", "evidence": "Table 1Association with overall survival and clinicopathologic characteristic in TCGA patients using Cox regression analysisClinical characteristicsHR (95% CI)p-valueAge≤ 60Ref> 602.94 [1.14–2.89]0.012*StageStage IRefStage II2.62 [0.82–3.59]0.154Stage III4.55 [1.96–5.17]< 0.001***Stage IV12.39 [4.69–15.39]< 0.001***NR3C1LowRefHigh2.17 [1.29–3.04]0.00173***", "target_result": "The overall-survival hazard ratio for high versus low NR3C1 expression in UCEC was 2.17.", "number_stated": true, "source_locator": "chars 11895-12253 of source sha256 a6ce82b9bd8746c2f3aeb18f12f6770546e08f8897c46ec67227ced4334bde7c"}, "recorded_at": "2026-10-31T15:36:33.963035+00:00"}], "agreement": false, "disagreements": {"target_result": {"minimax": "NR3C1 had an ROC AUC of 0.914 in GSE17025, 0.791 in GSE36389, 0.786 in GSE63678, and 0.882 in GSE115810; all four AUC values exceeded 0.78, indicating substantial diagnostic sensitivity and specificity for UCEC.", "minimax_alt": "High versus low NR3C1 expression was associated with overall survival in UCEC (HR 2.17, 95% CI 1.29–3.04, p = 0.00173), and NR3C1 was reported as an independent prognostic factor.", "decisive": false}, "analysis_methods": {"minimax": ["Receiver Operating Characteristic (ROC) curve analysis", "area under the ROC curve (AUC)", "pROC package in R", "Decision Curve Analysis (DCA)"], "minimax_alt": ["Cox proportional hazards regression analysis"], "decisive": false}, "statistical_method": {"minimax": "Receiver Operating Characteristic (ROC) curve analysis", "minimax_alt": "Cox proportional hazards regression analysis", "decisive": true}}, "agreement_basis": {"policy": "decisive_fields", "decisive_fields": ["work_type", "number_stated", "parent_value", "unit"]}, "blinded": true}, "human_critique": {"status": "not_run", "reviewer": null, "signed_at": null, "conflict_checked": false, "commentary": null}, "stages": [{"key": "target_materials", "status": "pending", "reason": null, "evidence": null}, {"key": "independent_readings", "status": "blocked", "reason": "The independent readings are incomplete or disagree; the source evidence must resolve the difference.", "evidence": {"provider_errors": []}}, {"key": "source_resolution", "status": "pending", "reason": null, "evidence": null}, {"key": "protocol_reconstruction", "status": "pending", "reason": null, "evidence": null}, {"key": "control_execution", "status": "pending", "reason": null, "evidence": null}, {"key": "control_comparison", "status": "pending", "reason": null, "evidence": null}, {"key": "publication", "status": "pending", "reason": null, "evidence": null}], "gaps": ["review.human_critique", "review.machine_critique.blocked"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "package:5ea97692b78bb4d3c2450962470fb00ab221923a7522691c8db47295f70021e4", "type": "sep_package", "label": "SEP v1 package for pmid:40113641"}], "activities": [{"id": "activity:conversion:pmid:40113641", "type": "conversion", "used": ["package:5ea97692b78bb4d3c2450962470fb00ab221923a7522691c8db47295f70021e4"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:minimax", "type": "model", "label": "minimax"}, {"id": "agent:minimax_alt", "type": "model", "label": "minimax_alt"}, {"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}], "source_run_id": "run-20261023T122449232913", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "pmid:40113641", "parent_hash": "sha256:08418d9c952f65d026a9562075115a2022690f06a76001bab44edb8f4ba278a8", "fork_of": null, "generation": 2}, "gaps": ["abstract", "asset:source:pmid:40113641 not verified", "authors", "capability", "corpus_record_absent", "execution.not_run", "method", "review.human_critique", "review.machine_critique.blocked", "step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40113641' does not resolve to a declared asset", "targets"]},"meta":{"api_version":"v1","automation_ladder_snapshot":"ba34b49490d9bfaf","automation_ladder_caveats":["Method extraction is rule-based and has no hand-judged accuracy sample. Evidence bands order textual support; method agreement intervals are null (n=0), never probabilities. Work-type validation does not validate methods.","Operation coverage describes only explicitly named methods. It does not prove a complete protocol, compatible parameters, successful compilation or re-execution. Unrecognised or ambiguous methods require review.","Data reachability is a textual accession or public-data statement. No dataset was downloaded or checked for access, completeness or licence.","Capabilities are planning states seeded from the dated equipment register. No bench ownership, completed quote or partner booking is inferred. Procurement counts are dependencies, not guaranteed replications or financial returns.","Tier and blocker counts are deterministic assessments of this corpus, not measured agreement. R3 is reserved for procurement; unresolved methods and data require human work (R4). The most restrictive axis wins, with capability, method, data as tie-breaks. Resolving one blocker can leave another at the same tier."],"engine":"corpus_engine","engine_provider":"rule_based_v1+lex99e1195c7d6f","run_id":"run-20261023T122449232913","source":"esearch","source_detail":"PubMed mirror biomed_pubmed (read-only)","run_status":"complete","imported_at":"2026-11-03T20:16:52+00:00","generated_at":"2026-11-04T02:12:26+00:00","total_papers":5529,"total_assays_tracked":29,"automatable_assay_families":6,"validated":true,"caveat_count":9,"blocking_caveats":5,"caveats":[{"code":"low_evidence_dominates","basis":["4,152 of 5,529 classifications (75.1%) matched no cue evidence and sit at the 0.05 floor.","34 of 230 data_only calls (14.8%) are in that group."],"claim":"Work type and confidence bands","gloss":"Most records were classified from their title and framing alone, with assay-cue coverage at zero.","detail":"Most classifications in this run rest on no matched evidence at all. At the evidence floor the work type comes from title framing and a fixed prior, not from anything in the paper, so the published work-type split should be read as a rough triage signal rather than a measured distribution. The impact is bounded but not removed: since 2026-10-23 the engine excludes floor-confidence records from the shortlist entirely and publishes them separately as held out, so the recommendation is unaffected even though the census is.","status":"unverified","severity":"high"},{"code":"runtime_is_modelled","basis":["Runtime is summed from published per-assay hour figures. Timing awaits the first executed replication.","4,407 of 5,529 records (79.7%) model to 0.00 hours because no assay was matched, which the ranking then reads as \"cheapest\".","0 of 10 shortlisted records carry a 0.00-hour estimate."],"claim":"Estimated replication runtime","gloss":"The hour figures are estimates built from assay mentions; timing awaits the first executed replication.","detail":"The hour figures are a cost model derived from the bench steps implied by each matched assay family. The first executed replication will supply measured timing. Records without an assay match currently receive a zero-hour model value.","status":"unverified","severity":"high"},{"code":"signal_score_without_citations","basis":["5,529 of 5,529 records (100.0%) have no citation count from the source metadata."],"claim":"Signal score","gloss":"The signal score uses publication-type weight because the mirror supplied zero citation counts.","detail":"The source supplied no citation counts for any record in this run, so the signal score is publication-type weight only. The citation term of the ranking is inert here, and the shortlist is ordered by publication type alone.","status":"unverified","severity":"high"},{"code":"abstract_absent","basis":["1,667 of 5,529 records (30.2%) carry no abstract at all."],"claim":"Evidence base for classification","gloss":"Some records contain a title and publication type only, defining the available classification evidence.","detail":"Where the mirror contains a title and publication types only, those fields define the classifier evidence. A methods section is required for fuller automated and human review.","status":"unverified","severity":"medium"},{"code":"classifier_agreement_measured","basis":["Published two-way decision agreement 77.7% (310 of 399 decidable records) from a stratified hand-check of 438 real records.","Three-way diagnostic agreement 73.9% (288 of 390 decidable records).","data_only agreement 75.2% (100 of 133 decidable).","48 record(s) could not be decided from title and abstract and were excluded from the denominator.","56 analysis-only records were published under another work type, so the data_only share is not an upper bound."],"claim":"Published automatable decision","gloss":"This figure has been checked against a human reading the same papers, so we know how often it is right.","detail":"The published automatable / not-automatable distribution comes from a rule-based classifier that was hand-checked against a fixed ground-truth definition. The agreement rates below are measured, not estimated. The three-way work type remains a diagnostic and is reported separately.","status":"measured","severity":"high"},{"code":"llm_comparison_not_measured","basis":["rule_based_v1+lex99e1195c7d6f produced 5,529 of 5,529 classifications.","Measured on 438 labelled records; these are RULE-BASED figures, not model results."],"claim":"Rule-based versus LLM agreement","gloss":"Every number here is the rule engine’s own. Model comparison begins when a working API credential is connected and measured.","detail":"The published two-way decision agreed with a human reading on 310 of 399 decidable records (77.7%). The diagnostic three-way label agreed on 288 of 390 decidable records (73.9%). Every percentage in this run belongs to the rule engine. Model comparison begins when a working API credential is connected and measured.","status":"not_measured","severity":"high"},{"code":"confidence_is_a_band","basis":["low band: 135 of 210 hand-judged records were correct, 64.3% [57.6, 70.5] (95% CI).","medium band: 80 of 104 hand-judged records were correct, 76.9% [68.0, 84.0] (95% CI).","high band: 73 of 76 hand-judged records were correct, 96.0% [89.0, 98.7] (95% CI).","A fitted calibrated probability was built, measured and withdrawn: it overstated held-out accuracy and reorders nothing."],"claim":"Confidence band and evidence score","gloss":"The confidence band carries a measured accuracy interval; 0.85 represents relative rank strength.","detail":"Confidence on this run is a BAND plus its measured accuracy interval. The number stored alongside each record is a monotone evidence score that orders records; 0.85 represents relative rank strength. A fitted probability was withdrawn after it overstated held-out accuracy. Threshold on the band.","status":"measured","severity":"medium"},{"code":"three_way_split_is_diagnostic","basis":["The corpus is published at the two-way decision: automatable or not-automatable. The three-way split (data_only / physical / mixed) is published as a diagnostic with its measured reliability attached, not as a claim, because the rule engine's `mixed` call cannot be defended.","The engine called 22 record(s) mixed; 10 matched the human judgement (45.5%, 95% CI 26.9-65.3).","The engine's `mixed` call is right on 10/22 hand-judged records (45.5%, 95% CI 26.9%-65.3%); a constant `physical` guess scores 36.4%.","The engine's `mixed` call does not reliably beat a constant `physical` guess on the hand-judged records it labels mixed.","The sign of that difference flips between two independent samples, so the three-way refinement is not demonstrably informative.","The mixed stratum contributes a disproportionate share of errors relative to its size.","At title-only depth the physical/mixed split collapses: a title says that work was done, not whether it was also analysed."],"claim":"The published decision is two-way","gloss":"The three-way split is shown for diagnosis only. The published split is two ways.","detail":"Count and filter the corpus on automatable versus not-automatable. The data_only / physical / mixed split remains an internal engine diagnostic because the mixed call has lower measured reliability.","status":"measured","severity":"medium"},{"code":"automatable_flag_unproven","basis":["6 of 29 assay families are marked automatable now: meta_analysis, machine_learning, survival_analysis, pca, molecular_docking, differential_expression.","The flag is a rule over assay family and work type; validation begins with the first executed replication.","It inherits the 75.2% data_only agreement rate, because the flag requires the record to be analysis-only."],"claim":"Automatable now","gloss":"This flag marks assay families the rules classify as data-only. The first executed replication will validate it.","detail":"This flag marks families the rules consider replicable from data alone. Its validity depends on the analysis-only classification and will be measured through an executed replication.","status":"unverified","severity":"medium"}],"confidence_policy":"band_with_measured_interval","confidence_contract":{"policy":"band_with_measured_interval","summary":"Confidence is reported as a BAND with a measured accuracy interval. The raw evidence score supplies monotone ordering. A fitted point estimate was withdrawn after it overstated held-out accuracy.","evidence_score":"A monotone evidence score in [0,1] built from cue mass, axis separation and text coverage. Use it to order records. Interpret 0.85 as relative rank strength.","confidence_band":"low | medium | high. Consumers threshold on this value.","band_accuracy":"Wilson 95% interval on the measured agreement of the hand-judged records in this band.","calibrated_confidence":"WITHDRAWN. Reintroduction requires held-out evidence demonstrating reliable estimates across every bin."},"band_accuracy":{"low":{"n":210,"low":0.576,"band":"low","high":0.7046,"agree":135,"accuracy":0.6429},"high":{"n":76,"low":0.8903,"band":"high","high":0.9865,"agree":73,"accuracy":0.9605},"medium":{"n":104,"low":0.6795,"band":"medium","high":0.8397,"agree":80,"accuracy":0.7692}},"provider":"rule_based_v1+lex99e1195c7d6f","lexicon_id":"99e1195c7d6f","contract_published_at":"2026-10-23T12:25:29.022232+00:00","confidence_is_band":true,"confidence_is_probability":false,"decision_taxonomy":"two_way","decidable_fraction_two_way":0.8854,"decidable_fraction_two_way_low":0.8565,"decidable_fraction_two_way_high":0.9143,"decidable_fraction_three_way":0.8586,"decidable_fraction_three_way_low":0.8287,"decidable_fraction_three_way_high":0.8884,"decidable_population":5529,"validation_sample_size":438,"three_way_cost_fraction":0.0268,"three_way_cost_records":148,"decision_counts":{"automatable":230,"not_automatable":5299},"decision_split":{"automatable":{"count":230,"share":0.04159884246699222},"not_automatable":{"count":5299,"share":0.9584011575330078}},"diagnostic_work_type_counts":{"mixed":24,"physical":5275,"data_only":230}}}